Prakash Chandra Sahu, Ramesh Chandra Prusty and Sidhartha Panda
The paper has proposed to implement gray wolf optimization (GWO)-based filter-type proportional derivative with (FPD) plus (1+ proportional integral) multistage controller in a…
Abstract
Purpose
The paper has proposed to implement gray wolf optimization (GWO)-based filter-type proportional derivative with (FPD) plus (1+ proportional integral) multistage controller in a three-area integrated source-type interlinked power network for achieving automatic generation control.
Design/methodology/approach
For analysis, a three area interconnected power system of which each area comprises three different generating units where thermal and hydro system as common. Micro sources like wind generator, diesel generator and gas unit are integrated with area1, area2 and area3 respectively. For realization of system nonlinearity some physical constraints like generation rate constraint, governor dead band and boiler dynamics are effected in the system.
Findings
The supremacy of multistage controller structure over simple proportional integral (PI), proportional integral, derivative (PID) and GWO technique over genetic algorithm, differential evolution techniques has been demonstrated. A comparison is made on performances of different controllers and sensitivity analysis on settling times, overshoots and undershoots of different dynamic responses of system as well as integral based error criteria subsequent a step load perturbation (SLP). Finally, sensitive analysis has been analyzed by varying size of SLP and network parameters in range ±50 per cent from its nominal value.
Originality/value
Design and implementation of a robust FPD plus (1 + PI) controller for AGC of nonlinear power system. The gains of the proposed controller are optimized by the application of GWO algorithm. An investigation has been done on the dynamic performances of the suggested system by conducting a comparative analysis with conventional PID controller tuned by various optimization techniques to verify its supremacy. Establishment of the robustness and sensitiveness of the controller by varying the size and position of the SLP, varying the loading of the system randomly and varying the time constants of the system.
Details
Keywords
Sonalika Mishra, Suchismita Patel, Ramesh Chandra Prusty and Sidhartha Panda
This paper aims to implement a maiden methodology for load frequency control of an AC multi micro-grid (MG) by using hybrid fractional order fuzzy PID (FOFPID) controller and…
Abstract
Purpose
This paper aims to implement a maiden methodology for load frequency control of an AC multi micro-grid (MG) by using hybrid fractional order fuzzy PID (FOFPID) controller and linear quadratic Gaussian (LQG).
Design/methodology/approach
The multi MG system considered is consisting of photovoltaic, wind turbine and a synchronous generator. Different energy storage devices i.e. battery energy storage system and flywheel energy storage system are also integrated to the system. The renewable energy sources suffer from uncertainty and fluctuation from their nominal values, which results in fluctuation of system frequency. Inspired by this difficulty in MG control, this research paper proposes a hybridized FOFPID and LQG controller under random and stochastic environments. Again to confer viability of proposed controller its performances are compared with PID, fuzzy PID and fuzzy PID-LQG controllers. A comparative study among all implemented techniques i.e. proposed multi-verse optimization (MVO) algorithm, particle swarm optimization and genetic algorithm has been done to justify the supremacy of MVO algorithm. To check the robustness of the controller sensitivity analysis is done.
Findings
The merged concept of fractional calculus and state feedback theory is found to be efficient. The designed controller is found to be capable of rejecting the effect of disturbances present in the system.
Originality/value
From the study, the authors observed that the proposed hybrid FOPID and LQG controller is robust hence, there is no need to reset the controller parameters with a large change in network parameters.